miRDB_McGuireSams_GradProjec-BIOL6860
Cierla McGuire Sams
3/6/2022
miRDB: miRNA binding site prediction software
miRDB is a great tool for predicting the binding sites of miRNAs in a transcript, and especially in the 3’ UTR.
All that is needed for using this tool is a computer with WiFi.
1. Go to http://mirdb.org/
Select the desired function. On the home page, one can search for a specific miRNA or Gene Target
For this tutorial we will focus on “Custom Prediction”.
### Select Custom Prediction from the list of functions on the left. 
### 2. For this tutorial, we will select ‘Human’for Species and the ’Sample target mRNA sequence’ as an example, which will auto-populate when selected. ### Importantly, if you input your own sequence from UCSC or anywhere else, use FASTA format. ### This software is forgiving of spaces and numbers input with the FASTA sequence, however,t his program does not allow you to utilise the carat ‘> NM_0000 Sample’ format at the beginning, so leave this out. ### Select ‘Go’ to initiate the prediction.
### 3. When it is finished running, select ‘Retrieve Prediction Results’
### 4. The output file will describe how many predicted miRNAs target the input region. ### THe predicted miRNAs will be ranked in order of Target Score. ### It is important to note that sometimes the outputs of similar sequences will vary a bit in the exact order of miRNAs listed for this reason–verify that the miRNA you compare is in both by examining the entire list–it may be in a different order and still be predicted to be present in the sequence.
### 5. One can click on a desired miRNA for information regarding the specific site locations by clicking the red hyperlink hsa-miR-5011-5p, for example. ### The higher the target score, the more one can trust the prediction. Prediction scores over 80 are thought to be valid. ### This also gives the exact miRNA Sequence, the Seed Location(s) and the input Target Length. ### When analyzing two separate inputs, such as a Wild-type sequence vs. a Mutant sequence with an introduced variation, one can easily compare the number of seed locations found in one vs. the other by comparing these side-by-side.
### 6. Going back to the original Prediction output table, one can instead select the ‘Details’ button to the left of the miRNA of interest. ### This will display useful information regarding both the mature and precursor miRNAs, their sequences, genomic location, and validated and predicted targets. ### It is also helpful to read the References on literature in which the miRNA of interest has been investigated.
### 7. From the ‘Details’ page above, one can click on ‘Predicted Targets’ to see the full list that the miRNA of interest may target. ### Again, these are ranked and given a Target Score to help select those that are most likely to be valid targets. 
### 8. For this example, we will select the first Predicted Target by selecting ‘Target Gene’. ### This details all of the Seed Locations within its 3’ UTR, length of the 3’ UTR, the exact miRNA sequence and Target Score. ### It also includes a Gene Description and a helpful ‘NCBI Gene ID’ hyperlink to further investigate the targeted gene.
### 9. Clicking on the ‘NCBI Gene ID’ hyperlink will open a new window with the gene page in the NCBI database. ### From here, one can select ‘FASTA’ directly under the bold heading in order to view the nucleotide sequence. 
10. After selecting the FASTA sequence, the accession number for the target gene can be copied.
### 11. In a new browser window, search for ‘UCSC Genome Browser’ and select ‘UCSC Genome Browser Home’
### 12. Then, select ‘Genome Browser’ option (highlighted in green). 
### 13. From here, select ‘Human’ and select ‘Dec. 2013 GrCh38(hg38)’ ### Past the accession number obtained from the NCBI FASTA page, and select ‘Go’. 
### 14. Select the Dark blue ‘NCBI RefSeq genes’ track (highlighted in green).
### 15. This will display the RefSeq gene information for the target gene. ### Select the ‘mRNA/Genomic Alignment’ (highlighted in green).
### 16. This displays both the cDNA and genomic DNA sequences. This is where one can obtain the exact sequence regions of interest for the target gene.
### 17. We originally saw that the miRNA had binding sites in the 3’ UTR of the target gene, but using this cDNA sequence, we can also determine if it binds in the 3’ UTR (or the entire transcript– we will focus on the 5’ UTR for now in this tutorial). ### Copy the entire 5’ UTR (highlighted in green). ### Red letters at the beginning of the sequence denote non-coding regions; blue denote coding regions.
### 18. Open a new browser window with http://mirdb.org/ ### Paste the entire copied sequence into the text box. Select ‘Human’ for Species. This time, do NOT select Sample target mRNA input. Select ‘Go’. 
### 19. Again, select the Predicted Sequences output link. ### Now, one can see that there are two predicted miRNAs binding to the 5’ UTR of the target gene, and its nucleotide length.
### 20. From here, one can further investigate the miRNAs by both clicking on the red hyperlink with its name and/or selecting ‘Details’ to view the References and locations as we did before at the beginning of this tutorial.